Extreme PCR Meets High-Speed Melting: A Step Closer to Molecular Diagnostics “While You Wait”

The race for fast molecular diagnostics has been ongoing for several years. And for good reason. The ability to diagnose or monitor diseases at the point of care is increasingly important to population health, chronic disease, and prevention of acute hospital admissions and readmissions. It also provides clinicians access to rapid and actionable diagnostic results (1).

Although protein biomarker-based tests no doubt played, and will continue to play, a major role, the rise of DNA and RNA-based diagnostics in view of the superb sensitivity and specificity that can be obtained is unquestionable. Speed, however, has not been the strongest point in DNA diagnostics, because in most approaches a DNA amplification step is involved, which adds to an already time-consuming sample preparation step, required for effective PCR and genotyping. Accordingly, applications like molecular testing of residual tumor by monitoring mutations in surgical margins during operation have not taken off, because even a 15-min PCR may weigh heavily in this setting (2).

During the past 3 years, the DNA amplification puzzle has started to yield a solution. In 2015 Carl Wittwer and his group at the University of Utah came up with Extreme PCR, a PCR adaptation that achieves DNA amplification of short targets in just a few seconds (3). Extreme PCR adds to a long track record of major PCR developments from the Wittwer laboratory that include rapid-cycle PCR (4), the Lightcycler instrument (5), and high-resolution melting (HRM)² (6).

Extreme PCR, which is distinct from a commercial trademark known as xPCR, achieves specific amplification against our basic PCR instincts by including 10- to 20-fold higher amounts of primers and polymerase in the reaction. These amounts are something that no serious PCR textbook would recommend because too much polymerase or primer in standard PCR reactions often leads to nonspecific amplification. Yet, retrospectively, it all makes sense if one considers that the increase in reagent concentration is balanced by the ultrafast thermocycling speed. As the primer and polymerase concentrations are increased to enable faster cycling, specificity is compromised, but this is balanced by less time for incorrect annealing (primer dimer or off target) (3). Thus, if you run regular or rapid-cycle PCR at extreme concentrations, you get junk. If you run regular or rapid-cycle concentrations at extreme speeds, you get no (or very low efficiency) product. If you match the increase in polymerase and primer concentrations (lowers specificity), to faster PCR (increases specificity), you get Goldilocks (3). Following this development, additional reports were published for achieving PCR in under 1–5 min via heating on nanoparticle or microfluidic platforms (7), with Extreme PCR still achieving the shortest amplification times.

In another counterintuitive development, the Wittwer laboratory showed that genotyping a PCR product via high-speed melting (HSM, a fast-melting version of the established HRM) improved genotyping of short (<100 bp) amplicons when performed at DNA melting speed rates 8–16 °C/s (8). This speed is 1–2 orders of magnitude faster melting than the gradual heating at 0.01–1 °C/s used for standard HRM. And while this unexpected observation is not yet fully understood, it seems to be partly due to restricting the time that heteroduplexed amplicons have available for reverting to homoduplexes via recombination, which reduces the discrimination in standard HRM (8). This explains the better heteroduplex discrimination observed for HSM.

This issue of Clinical Chemistry includes the latest addition in the Wittwer suite of PCR developments (9), filling in another crucial piece of the rapid molecular diagnostics puzzle: the instrumentation. By optimizing the temperature control loop feedback for fast temperature control on a microfluidic device, Extreme PCR is combined in series with HSM, yielding amplification and genotyping from hepatitis B virus and Clostridium difficile in just 52–87 s. Although the quantitative PCR and temperature precision aspects still need to be improved for microfluidic Extreme PCR-HSM, the data indicate that, as an end point approach, the application already seems robust.

By overcoming the amplification, genotyping, and instrumentation issues, this last development completes one of the remaining steps to bring PCR-based molecular diagnostics closer to point-of-care application. Yet, if one were to roll back time by a few years, few experts would have predicted that the first manifestation of under 1-min amplification and genotyping would come from PCR. Why thermocycle a reaction at high temperatures when isothermal amplification can do this with trivial instrumentation and at constant, near-ambient temperatures? Indeed, exponential isothermal amplification using nucleic acid sequence-based amplification, loop-mediated isothermal amplification, recombinant polymerase amplification, helicase dependent amplification, rolling-circle amplification, and strand displacement amplification circumvent the need for a thermocycler while also displaying tolerance to less-than-perfect DNA extraction methods (10). But, for the time being none of the isothermal methods displays robust and reproducible amplification times <5–10 min, even without counting the genotyping step. This advancement serves as another reminder that disruptive innovation can always overturn expectations.

By introducing annealing and thermocycling time as a major variable in PCR, Extreme PCR coupled with HSM and microfluidics opens the possibility for further innovations. Attempting to look at the crystal ball in terms of impact on mutation detection technologies, fast coamplification at lower denaturation temperature (fast COLD-PCR) employs a 2-step cycling just like Extreme PCR while using denaturation at lower temperatures around 80–87 °C (11). The reduced temperature ramping requirement is likely to enable extreme fast-COLD-PCR to operate at even faster rates than Extreme PCR, while also enabling mutation enrichment and prominent HRM melting peaks for low-level mutations or the possibility of sequencing the final product (12). Multiplexed PCR is another application that may get a boost under Extreme PCR conditions because most nonspecific interactions (primer to primer, or primer to wrong template) involve mismatched hybridizations that require more time to form than hybridization of fully matched template. Given that time is a tightly controlled variable in Extreme PCR, it is plausible that multiplexed Extreme PCR may present less mispriming. Combined with multiplexed HRM genotyping (13) or multiplexed HRM scanning (14), this reduction may enable advanced forms of this technology. Similarly, allele-specific amplification (15) also relies on fully matched vs 3′ end mismatched hybridization. Hence, with appropriate optimization, one might be able to take advantage of the limited time provided to a polymerase for extending a fully matched vs mismatched template and thereby increasing the allele-specific amplification genotype discrimination ability. On the other hand, the combination of Extreme PCR with Taqman or molecular beacon genotyping would use probes in the path of the polymerase and likely slow down the polymerase and lengthen amplification times. Conversely, fluorescence ratio-based methods yield genotypes without requiring a melting step and without reactions needing to reach full cycles, thus dialing back down overall times needed for amplification plus genotyping. The net effect on reaction speed with fluorescent probes remains to be explored. As always, with all these aspirations the devil is in the details. The repeatability and robustness of each approach will ultimately determine its utility.

Nevertheless, now that amplification and genotyping speed have increased, the initial sample preparation and DNA and RNA extraction step that inevitably precedes PCR still stand in the way. Is extreme sample extraction the next development waiting to happen? Multiple efforts are indeed ongoing in this direction (16). It appears that, as the pieces of the puzzle gradually come together, molecular diagnostics “while you wait” will become a reality sooner rather than later.